{"as_of":"2026-08-08T12:44:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:40525d8197e8ef38885969214a8540b3ceb7e1a7120c5210e5dade470b7d8e3a","coverage":[{"denominator":19,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":19,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-02T20:47:52.010550Z","state":"measured"},{"denominator":19,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":19,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-08T06:32:00.761636+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2602.22352/citation-record","integrity":"/paper/2602.22352/integrity","json":"/paper/2602.22352/citation-record.json","paper":"/paper/2602.22352"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.521086Z","title":"FINN: A framework for fast, scalable binarized neural network inference","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.521086Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:97648721057bea90f11bea6c400f933387eb3837bb2591fa2b7f432d86610e94","observation_id":"149b56eb-febf-4e38-a660-8b45d2499c53","resolution":{"observed_at":"2026-08-02T20:47:50.521086Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.651140Z","title":"FINN-R: An end-to-end deep-learning frame- work for fast exploration of quantized neural networks","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.651140Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:1ed94acfcfe2a1dfdc7cfb53cd473d4429df0bf2af2e19fe76e7768a8405320f","observation_id":"d59770c5-b490-4668-b202-6a70ac3a57bf","resolution":{"observed_at":"2026-08-02T20:47:50.651140Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.745626Z","title":"Bitwise Systolic Array Architecture for Runtime-Reconfigurable Multi-Precision Quantized Multiplication on Hardware Accelerators","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.745626Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:a9daabacaf52a9c7c3a441064349a712d2db0e9373acc40e2234d94ca3bf0ae2","observation_id":"6b192046-ac01-41c6-a1a2-1fda3e92ec40","resolution":{"observed_at":"2026-08-02T20:47:50.745626Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.821633Z","title":"The mnist database of handwritten digit images for machine learning research","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.821633Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:5fe81c1c3c006038ebe35fbff5d55aeeb9cca0d2a69bd186a19194f59a31d1b4","observation_id":"a53eef20-5c48-445e-9276-55cbea287cd5","resolution":{"observed_at":"2026-08-02T20:47:50.821633Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.889534Z","title":"Sigmoid-weighted lin- ear units for neural network function approximation in reinforcement learning","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.889534Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:ea6ecf9a46eef07befd61e03fde93867ef26b16b64da5c3f138fb4d25b51e49f","observation_id":"55a50913-55e7-45db-87e7-68861ebf994c","resolution":{"observed_at":"2026-08-02T20:47:50.889534Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:50.969543Z","title":"Neural networks with digital LUT activation functions","venue":null,"work_id":null,"year":1993},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:50.969543Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:9fa64d1f40cb3dc3914fbef81564dfbb573c2a096e8addc09f3d4e7667a6d9fa","observation_id":"8c4e25d1-5fad-4734-91fb-244782d18a32","resolution":{"observed_at":"2026-08-02T20:47:50.969543Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.053837Z","title":"Design and FPGA Implementation of the LUT based Sigmoid Function for DNN Applications","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.053837Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:cf33a6a33174a1e5ab0535c8e22b46bdf9c1d392f2475d68b1e08e2659680f5b","observation_id":"89aaa6c3-135a-4d8a-8adc-4442f4c19e5d","resolution":{"observed_at":"2026-08-02T20:47:51.053837Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.121950Z","title":"An optimized lookup-table for the evaluation of sigmoid function for artificial neural networks","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.121950Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:3fb4a3d61b1f436c3cf6cb5212f0e704b94f7b8fe0bb229c731436c1dfa86168","observation_id":"9c1d8a3f-de35-405a-bcdb-7d6d63f901aa","resolution":{"observed_at":"2026-08-02T20:47:51.121950Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.202564Z","title":"Sig- moid generators for neural computing using piecewise approxima- tions","venue":null,"work_id":null,"year":1996},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.202564Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:6edc2d686ceb41f471508db38379a0b0297fa9a6613051587721679def98ccb2","observation_id":"82a87ef1-702c-4f0f-8674-c5e7dc569f41","resolution":{"observed_at":"2026-08-02T20:47:51.202564Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.296328Z","title":"FPGA Implementation for the Sigmoid with Piecewise Linear Fitting Method Based on Curvature Analysis","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.296328Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:77ee91c5047a82986fc4c7a66dfe90b6ee82481d8d6026a74186f65e29e341a9","observation_id":"55060a08-6197-4962-9a2a-255ebe8460e6","resolution":{"observed_at":"2026-08-02T20:47:51.296328Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.395155Z","title":"Hardware Implementation of Sigmoid Activation Functions using FPGA","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.395155Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:623795903ccf463786f77917bb9255b55a1a95050cdcad97ac80a3384d831775","observation_id":"962234c1-d7e5-47ed-a129-e9edc6d1485e","resolution":{"observed_at":"2026-08-02T20:47:51.395155Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.484940Z","title":"Low Complexity Sigmoid Function Implementation Using Probability-Based Piecewise Linear Function","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.484940Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:3f0ecd777a0752123a854876bf05c651163f2068e61b0687777695aa9fac1894","observation_id":"1b6f41f7-b752-4d46-9861-67ec8d485f78","resolution":{"observed_at":"2026-08-02T20:47:51.484940Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.544666Z","title":"Cost effective Tanh activation function circuits based on fast piecewise linear logic","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.544666Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:74b91cee2b64aa07809b0e99ea9e3d8568d04f8c53c37692f937c5831c389b86","observation_id":"afa220af-5b99-41c5-a32c-c8e599a34465","resolution":{"observed_at":"2026-08-02T20:47:51.544666Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.623183Z","title":"Hardware Implementation of Tanh Exponential Activation Function using FPGA","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.623183Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:3ca881c0ad4629d394937537d992c8a9336a36ccc828549dc2574c6787c879c8","observation_id":"c0e5eb6c-e04e-4574-b4b7-46a5cf60e6c6","resolution":{"observed_at":"2026-08-02T20:47:51.623183Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.05025","last_updated":"2022-03-09T19:57:14Z","snapshot_observed_at":"2026-07-06T12:46:12.270140Z","submitted_at":"2022-03-09T19:57:14Z","title":"Power-of-Two Quantization for Low Bitwidth and Hardware Compliant Neural Networks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.05025","snapshot_observed_at":"2026-08-02T20:47:51.706464Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.706464Z"},"links":{"cited_paper":"/paper/2203.05025","citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:201bc2b52eb5f9ec98a6f869a2648f8e7489e223acc17ee7dad8af63c5e71a3a","observation_id":"8acb6229-e923-4a9f-9c3c-873ec0b23145","resolution":{"observed_at":"2026-08-02T20:47:51.706464Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1909.13144","last_updated":"2020-02-02T14:15:07Z","snapshot_observed_at":"2026-07-06T08:25:25.438850Z","submitted_at":"2019-09-28T20:14:11Z","title":"Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1909.13144","snapshot_observed_at":"2026-08-02T20:47:51.784991Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.784991Z"},"links":{"cited_paper":"/paper/1909.13144","citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:9f335059e068cac63b3cc46e717fce9ff4013bd0c40bbc86590b2149ec28af82","observation_id":"3290b408-9e7d-4521-8bb7-de67f77074fd","resolution":{"observed_at":"2026-08-02T20:47:51.784991Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.868397Z","title":"Jekel and Gerhard Venter.pwlf: A Python Library for Fitting 1D Continuous Piecewise Linear Functions","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.868397Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:a5835cda56245161e5d18327c406228771ff3c5103ac375c0d9fdf32fc9c2f2b","observation_id":"395c6437-a5b3-46d2-9283-7f4ec93e00d7","resolution":{"observed_at":"2026-08-02T20:47:51.868397Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:51.951548Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:51.951548Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:9c214ea4840002406aa1ef6550dac3de3c39095c69fdbcad69c0afd9fb58d32b","observation_id":"2b6d6c4a-a604-4015-919d-696d19f0e145","resolution":{"observed_at":"2026-08-02T20:47:51.951548Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-02T20:47:52.010550Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-02T20:47:52.010550Z"},"links":{"citing_paper":"/paper/2602.22352"},"observation_digest":"sha256:3337215d8622b41e0110e3fb574f0904c54d6d1b8ef6e48ee508eafd69f7a8fa","observation_id":"a723fc4b-c60b-4df6-ac4d-dcd158c1e279","resolution":{"observed_at":"2026-08-02T20:47:52.010550Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2602.22352","last_updated":"2026-06-09T16:42:58Z","latest_version":2,"primary_category":"cs.AR","snapshot_observed_at":"2026-08-02T20:47:50.145461Z","submitted_at":"2026-02-25T19:18:22Z","title":"GRAU: Generic Reconfigurable Activation Unit Design for Neural Network Hardware Accelerators"},"reference_resolution":{"displayed":19,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":19,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":19},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"thesis":"As of 8 August 2026, this Paper Citation Record lists 19 of 19 outbound references and 0 inbound Pith citation observations for arXiv:2602.22352."}